CN105809279B

CN105809279B - A kind of water resources optimal operation method based on multi-target quantum shuffled frog leaping algorithm

Info

Publication number: CN105809279B
Application number: CN201610122222.XA
Authority: CN
Inventors: 郭玉雪; 方国华; 付晓敏; 闻昕; 袁玉
Original assignee: Hohai University HHU
Current assignee: Hohai University HHU
Priority date: 2016-03-03
Filing date: 2016-03-03
Publication date: 2019-05-31
Anticipated expiration: 2036-03-03
Also published as: CN105809279A

Abstract

The invention discloses a kind of water resources optimal operation methods based on multi-target quantum shuffled frog leaping algorithm, and steps are as follows: obtaining the essential information data of water resource system first；Next establishes Programming for Multiobjective Water Resources Optimal Operation Model；Then it executes and is based on multi-target quantum shuffled frog leaping algorithm, solve the Pareto optimal solution set of water resource system Multiobjective Optimal Operation；Finally according to Multi-Objective Decision Theory, subjective and objective weight, which combines, selects optimal water resource scheduling theory.The present invention realizes global optimizing, improves computational efficiency, meets selection water resource system multiobjective optimization scheduling scheme requirement.

Description

Water resource optimization scheduling method based on multi-target quantum mixed frog leaping algorithm

Technical Field

The invention belongs to the technical field of water resource scheduling in the field of water conservancy and hydropower, and particularly relates to a water resource optimization scheduling method based on a multi-target quantum mixed frog leaping algorithm.

Background

The water resource optimization scheduling is an optimal control problem of a multi-constraint multi-stage decision-making dynamic and complex nonlinear system, and particularly increases the complexity of optimization solution for multi-target comprehensive utilization projects such as flood control, power generation, irrigation, water supply, shipping, sand discharge and the like. The research of water resource optimization scheduling starts from the reservoir optimization scheduling problem proposed by Masse in the 40 th of the 20 th century, and the system engineering technology is widely applied to the water resource optimization scheduling in the middle of the 50 th. In recent years, along with the gradual improvement of mathematical programming theory and the wide application of computer technology, the technology and the method of water resource optimization scheduling are further enriched, and Evolutionary Algorithms (EA) such as genetic algorithm, fuzzy degradation, artificial neural network, chaos optimization algorithm, ant colony algorithm and the like are gradually and widely applied to the water resource optimization scheduling. In order to maximize the overall comprehensive benefit, the objectives need to be balanced and preferred, and many researchers are engaged in the research of the multi-objective evolutionary algorithm (MOEA). Most of recent research on MOEA is to introduce a non-dominated solution set (Pareto) idea in a population evolution process, and a solid foundation is provided for multi-objective optimization scheduling of water resources.

Quantum computing is a new interdisciplinary subject formed by fusing quantum theory and information science, takes factorization quantum algorithm proposed by Shor in 1994 and random database search quantum algorithm proposed by Grover in 1996 as marks, and rapidly becomes a research hotspot due to excellent computing performance. In 1996, the quantum genetic derivation algorithm proposed by the british scholars narayana opened a new trend of combining quantum computing with evolutionary algorithms. The Quantum Evolution Algorithm (QEA) introduces quantum bit codes, has the characteristics of small population scale and strong global optimization capability, and has the capability of collaborative parallel search.

The mixed Frog-leap Algorithm (SFLA) is a post-heuristic computing technology based on group intelligence, is proposed by Eusuff and Lansey in 2003, has the characteristics of less Algorithm parameter setting and overlapping of local search and global search, and is gradually and successfully applied in the fields of pattern recognition, signal and information processing and function optimization. Similar to other intelligent optimization algorithms, basic SFLA also has the problems that the optimization capability of the algorithm depends on parameter setting, the later stage is easy to fall into the local optimal solution, the convergence speed is slow, and the like, and in the initialization stage of the SFLA, the distribution property of the initial population will influence the convergence performance of the whole algorithm.

In recent years, many scholars at home and abroad introduce quantum concepts into SFLAs, quantum mixed frog-leaping algorithms (QSFLAs) are proposed, and multi-target quantum mixed frog-leaping algorithms (MQSFLAs) are proposed by combining Pareto on the basis of the quantum mixed frog-leaping algorithms, so that the quantum mixed frog-leaping algorithms are widely applied to partial projects and disciplines, but are not applied to the field of water resource optimization scheduling at present, and the multi-target quantum mixed frog-leaping algorithms proposed at the present stage have the defects of poor initial population distribution, easy local optimal solution, low convergence speed and the like. Meanwhile, in the field of water resource optimization scheduling, when a multi-objective decision method is adopted to process a non-inferior solution set, evaluation index weight needs to be determined based on the combination of subjective and objective factors.

Disclosure of Invention

The purpose of the invention is as follows: aiming at the defects that the traditional SFLA is easy to fall into local optimal solution, the convergence speed is low and the like, and aiming at avoiding the influence of the initial population distribution difference on the optimization process, quantum computation is introduced into the SFLA, a water resource scheduling method based on a multi-target quantum mixed frog leap algorithm (MQSFLA) is provided, and a multi-target decision method based on the combination of main and objective weights is adopted to process a non-inferior solution set, so that the multi-target optimal scheduling of a reservoir is realized.

The technical scheme is as follows: a water resource optimization scheduling method based on a multi-target quantum mixed frog leaping algorithm comprises the following steps:

the method comprises the following steps: acquiring basic information data of a water resource system project, comprising the following steps: the overflow capacity value q of the pump, the gate station and the reservoir, the initial and final storage capacity limit V of the lake and the reservoir, and the normal water storage level Z of the lake and the reservoir_{Is just}Flood control water level limit Z_DefendDead water level Z_{Death by death}Lake and reservoir volume-water level relation curves S-Z, reservoir downstream water level-discharge flow relation curvesLine Z-Q, reservoir generator set output constraint value N and water inflow W;

step two: establishing a multi-objective optimization scheduling mathematical model taking the maximum comprehensive benefits of social benefits, economic benefits, ecological environments and the like as target functions and considering the constraint conditions of water balance, unit output, overflow capacity and the like;

step three: the method for executing the multi-target quantum mixed frog leaping algorithm comprises the following steps:

(1) determining a water resource scheduling period T by taking the last water storage capacity (last storage capacity) S, the downward discharge flow u or the engineering water flow q of a water resource engineering period as decision variables;

(2) and setting parameters. Determining an initial population scale G, a sub-population number N, a sub-population individual number M, a global iteration number MAXGEN, a sub-population iteration number K and an external archive set scale N which are formed by decision variables_EARotation angle operator [ delta ]_min,δ_max]Frequency of variation [ P ]_min,P_max]；

(3) Generating an initial solution population based on a quantum three-chain encoding scheme: the qubits belonging to a combination of successive amplitude variables theta andthe vector space can be defined by a point on the Bloch sphere embedded in a three-dimensional Cartesian coordinate systemG initial individuals are generated based thereon and divided into N sub-populations, each sub-population containing M individuals;

(4) the initial global iteration times are set to GEN (0), and an external archive set (EA) is assigned to an empty set;

(5) performing solution space conversion and calculating each objective function value of each individual in the sub-population, performing non-dominated sorting, updating EA according to a dynamic update mechanism, and randomly selecting a global optimal solution X_g,bDetermining an optimal solution atCorresponding amplitude theta on Bloch sphere_g,b、The operation flow of the external archive set with the dynamic update mechanism is as follows:

① judging the number of non-inferior solutions, if the number of non-inferior solutions is larger than N_EAIf so, go to step ②, otherwise, go to step ③;

② calculating crowding distance of each non-inferior solution, assigning infinite crowding distance to the boundary point to ensure entering next generation, sorting according to the crowding distance, deleting the individual with the minimum crowding distance, updating the crowding distance again until reaching the scale of the external archive set, stopping, and outputting the updated individual to EA;

③ if the number of non-inferior solution sets is less than the EA set scale, calculating the congestion distance of I, II individuals with non-inferior solution grade, and recording the average congestion distance as d₁、d₂Individuals in the deletion sequence level I, II with congestion distances lower than the average congestion distance;

④ reproduction and optimization of Elite individuals in EA, randomly selecting global optimum solution X_g,bDetermining theta_g,b、

⑤, carrying out dominant comparison between non-inferior solution newly generated by global iteration and elite solution in EA, replacing dominant solution in EA, completing updating of EA, and randomly selecting global optimal solution X_g,bDetermining theta_g,b、

(6) Randomly ordering the population individuals in each grade according to non-inferior grades, dividing N sub-populations after all individuals are mixed, counting the number of individuals of each sub-population by M, selecting the first individual and the Mth individual in the sub-populations as the optimal solution and the worst solution of the sub-populations, and marking as X_b、X_wDetermining theta_b、And theta_w、

(7) Local search updating, namely updating the worst solution in each sub-population based on quantum computation, and specifically comprising the following steps:

① setting the optimal individual X in the current sub-population_bAnd worst individual X_wJ (j ═ 1,2, …, T) quantum bits;

② determining the worst solution X in the current sub-population by adopting an individual updating strategy of rotating r around a solid axis_wTo the optimal solution X_bRotation updating, wherein a rotation angle operator is improved, and a rotation angle operator is dynamically adjusted according to the non-inferior sorting level;

③, executing individual variation operation, and calculating by adopting dynamic probability population diversity retention strategy;

④ calculating X_wAnd each objective function value of the variant individual, if the variant individual dominates X_wIf so, the variant individual replaces X_wOtherwise, replace X randomly_w；

⑤, when the iteration times of the sub-population reach K times, finishing the update iteration of the current sub-population, repeating the steps ① - ④, and performing the local search of the next sub-population;

(8) mixing the sub-populations, mixing all individuals after local search of each sub-population is finished, recombining the individuals into G individual populations, and turning to the step (5);

(9) and (4) judging whether the global iteration times MAXGEN are reached, if not, turning to the step (5) to continue the next round of global search, otherwise, finishing the algorithm and outputting EA.

Step four: based on EA, the optimal scheduling scheme of the water resource system is determined by adopting a multi-objective decision method based on combination weight, and the method comprises the following steps:

(1) and (3) determining subjective weight of the attribute by using an analytic hierarchy process: establishing a hierarchical structure of the system according to the selected target functions, performing pairwise comparison between indexes of all the hierarchies, establishing a judgment matrix, and calculating the weight w 'of each evaluation index'_j(j ═ 1, 2.., n), i.e., each evaluation protocol is described by n evaluation indices, and a consistency check is performed;

(2) determining the objective weight of the attribute by using an entropy weight method, which is specifically as follows:

① construct a relative membership matrix EA having N in common_EASet non-inferior solution, i.e. presence of N_EAScheduling scheme of water resource to be evaluated; calculating respective objective function values f of all non-inferior solutions_i,j(i＝1,2,…,N_EAJ 1,2,., n), calculating a relative goodness matrix R based on the index feature matrix normalization_i,j；

② calculating entropy of each evaluation index_jAnd its objective weight W_j；

(3) And (3) calculating and determining the attribute combination weight and the optimal scheduling result: when the subjective weight w 'of each evaluation index is obtained'_jAnd objective weight w ″_jThen, the combination weight w of the corresponding evaluation index can be obtained by calculation according to the preference coefficient l between the subjective and objective weights_jBased on the combined vector W ═ W₁,w₂,…,w_j]^TAnd multiplying the normalized decision matrix by the weight vector, and taking a scheduling scheme corresponding to the maximum value as an optimal scheme.

By adopting the technical scheme, the invention has the following beneficial effects:

(1) the requirement of multi-objective optimization scheduling of a water resource system is met;

(2) the initial population can exert the characteristic of quantum motion in space by using a quantum three-chain coding scheme, so that the population diversity is increased, and the problem optimization process is shortened, thereby avoiding the problem that the algorithm falls into local optimization due to the fact that the SFLA random initial population is concentrated in some local areas;

(3) an external archive collection method of a dynamic updating mechanism is adopted, so that the uniform distribution of non-inferior solution individuals is ensured, good diversity is realized, and the global convergence is accelerated;

(4) quantum computation is introduced into SFLA, a strategy of dynamically adjusting a rotation angle operator according to the individual non-inferior ranking level is provided, and population individuals can be promoted to converge to the non-inferior solution front edge as soon as possible; .

(5) The local updating mechanism of the SFLA algorithm is improved, a dynamic probability population diversity variation maintaining strategy is established, the population diversity is well maintained, meanwhile, the time for algorithm convergence is shortened, and the local searching capability of the SFLA is enhanced;

(6) the adopted multi-target decision method combines subjective and objective weights, enhances the autonomous decision capability of the multi-target scheduling system of the water resource system, and avoids excessive subjective preference brought in the decision process.

Drawings

FIG. 1 is a flow chart of a method of the present invention;

FIG. 2 is a flow chart of a multi-target quantum leapfrog algorithm;

FIG. 3 is a flow diagram of a multi-objective decision making method;

FIG. 4 is EA solution collection space distribution diagram of south-to-north water diversion east line engineering lake groups in open water;

FIG. 5 is a graph showing the change of the storage capacity of the east engineering lake group in the south-to-north water diversion in the open water year.

Detailed Description

The present invention is further illustrated by the following examples, which are intended to be purely exemplary and are not intended to limit the scope of the invention, as various equivalent modifications of the invention will occur to those skilled in the art upon reading the present disclosure and fall within the scope of the appended claims.

Aiming at the defects that the traditional SFLA has low convergence speed and falls into a local optimal solution and the like, the quantum computation is introduced into the SFLA by combining the characteristics of the quantum computation such as the traversability and the like, and the invention provides a water resource scheduling method based on a multi-target quantum leap algorithm (MQSFLA). The method generates an initial population by utilizing quantum three-chain coding, adopts an external filing collection method of a dynamic updating mechanism and introduces quantum computing to an SFLA local evolution search process, realizes the global optimization searching capability of the algorithm, and selects an optimal water resource scheduling scheme by adopting a multi-objective decision theory combining subjective and objective weights on the basis of a non-inferior solution set, thereby realizing the multi-objective optimal scheduling of the reservoir.

As shown in fig. 1, a water resource scheduling method based on a multi-target quantum mixed frog leaping algorithm mainly includes the following four steps:

the method comprises the following steps: acquiring basic information data of a water resource system project, comprising the following steps: the overflow capacity value q of the pump, the gate station and the reservoir, the initial and final storage capacity limit V of the lake and the reservoir, and the normal water storage level Z of the lake and the reservoir_{Is just}Flood control water level limit Z_DefendDead water level Z_{Death by death}Lake and reservoir volume-water level relation curves S-Z, reservoir downstream water level-discharge flow relation curves Z-Q, reservoir generator set output constraint value N and water input W;

minF(x)＝{f₁(x),f₂(x),…,f_n(x)} (1)

in the formula, N represents the target number of the optimized scheduling of the water resource system, and N is 1,2, …, N; f (x) -a set of objective functions; f. of_n(x) An objective function which is expressed as the maximum comprehensive benefit of social benefit, economic benefit, ecological environment and the like; Δ t — represents a calculation period interval; w is a_t,u_tM represents the inlet flow and the outlet flow of the units such as the reservoir, the pump station, the sluice and the like in the delta t period of the reservoir³/s；S_t,S_t+1Representing the water storage amount m at the end of t time period and t +1 time period of units such as reservoirs, pump stations, water gates and the like³；I_tRepresenting the amount of loss, m, over a period of deltat³；Z_t-represents the water level before the end of the time period t, m; z_t,min,Z_t,max-represents the minimum, maximum water level allowed at the end of the t period; q. q.s_t-representing the generated flow, m, over a period of deltat³/s；q_t,min,q_t,max-minimum and maximum let-down flow, m, allowed at the end of the t period³/s；N_tRepresenting the output of the hydropower station in the delta t period of the reservoir, kW; n is a radical of_t,min,N_t,max-representing the minimum and maximum allowable force output values, kW, at the end of the t period of the hydropower station; z, Z' -represents the initial and final water level limit value m of the reservoir.

Step three: a flow chart of the method for executing the multi-target quantum mixed frog leaping algorithm is shown in figure 2, and the method mainly comprises the following steps:

(1) determining a water resource scheduling period T by taking the last water storage capacity (last water level) S, the downward discharge flow u or the engineering overflow flow q of a water resource engineering period as decision variables;

(3) Generating an initial solution population based on a quantum three-chain encoding scheme: the qubits belonging to a combination of successive amplitude variables theta andthe vector space described can be described by a point on the Bloch sphere embedded in a three-dimensional cartesian coordinate system, and the ith individual can be represented as:

in the formula: theta is more than or equal to 0 and less than or equal to pi,i＝1,2,…,G。

based on this, G initial individuals are generated and divided into N sub-populations, each sub-population containing M individuals.

(5) performing solution space conversion and calculating each objective function value of each individual, performing non-dominated sorting, updating EA according to a dynamic update mechanism, and randomly selecting a global optimal solution X_g,bDetermining the theta corresponding to the optimal solution on the Bloch spherical surface_g,b、The operation flow of the external archive set with the dynamic update mechanism is as follows:

① setting the optimal individual X in the current sub-population_bThe j (j ═ 1,2, …, T) qubits are:

worst individual X_wThe j quantum bit is:

② determining the worst solution X in the current sub-population by adopting an individual updating strategy of rotating around a solid axis_wTo the optimal solution X_bRotation update, solid rotation axis: r ═ X_w×X_b(6)

ThenThe rotation matrix for a delta angle of rotation about axis r is as follows:

namely, it isOperated by an angle delta about axis rThe rotation angle operator is improved, and the rotation angle operator is dynamically adjusted according to the non-inferior sorting level, which is shown as the following formula:

in the formula: delta_w-representing the rotation angle of the worst individual of the current sub-population, R-representing the number of non-inferior solution ranking levels of the population, R_wIn-representation of sub-populationsNon-inferior solution ranking of worst individuals

③ performing individual mutation operations, the mutation strategy is as follows:

wherein,

wherein [ sigma ]_x,σ_y,σ_z]Into a set of pauli matrices

And (3) calculating by adopting a dynamic probability population diversity retention strategy, wherein the calculation formula is as follows:

in the formula, P_i-representing the ith global iteration frequency, P_i∈[P_min,P_max]

(9) judging whether the global iteration times MAXGEN are reached, if not, turning to the step (5), continuing the next round of global search, otherwise, finishing the algorithm and outputting EA;

step four: based on EA, determining an optimal water resource scheduling scheme by adopting a multi-objective decision-making method based on combination weight, wherein a flow chart of the method is shown in FIG. 3 and mainly comprises the following steps:

(1) and (3) determining subjective weight of the attribute by using an analytic hierarchy process: establishing a hierarchical structure of the system according to the selected target functions, performing pairwise comparison between indexes of all the hierarchies, establishing a judgment matrix, and calculating the weight w 'of each evaluation index'_j(j ═ 1, 2.., n), i.e., each evaluation protocol is described by n evaluation indices and a consistency check is performed;

① construct a relative membership matrix EA having N in common_EASet non-inferior solution, i.e. presence of N_EAScheduling scheme of water resource to be evaluated; calculating respective objective function values f of all non-inferior solutions_i,j(i＝1,2,…,N_EAJ 1,2,., n), calculating a relative goodness matrix R based on the index feature matrix normalization_i,j：

② calculating the entropy of each evaluation index, the calculation formula is as follows:

wherein,

③ calculating the objective weight of each index, it can be obtained by using the following formula:

(3) and (3) calculating and determining the attribute combination weight and the optimal scheduling result: when the subjective weight w 'of each evaluation index is obtained'_jAnd objective weight w ″_jThen, the combination weight of the corresponding evaluation index can be calculated by the following formula:

w_j＝lw′_j+(1-l)w″j (16)

wherein, l is a preference coefficient between the subjective weight and the objective weight, and l is (0, 1).

Based on the combined vector W ═ W₁,w₂,…,w_j]^TAnd multiplying the normalized decision matrix and the weight vector, and taking a scheduling scheme corresponding to the maximum value as an optimal scheme.

The multi-objective optimized scheduling of the lake group in the east line engineering of south-to-north water diversion is taken as an example to illustrate the effectiveness and rationality of the method. The south-to-north water transfer east line project is a giant water resource optimization scheduling system which spans two provinces and communicates rivers such as Yangtze river, Huaihe river, Shandong peninsula river, yellow river, sea river and the like, and relates to a giant water resource optimization scheduling system which has 1 hundred million and more water supply population, 69 planned water supply towns and 3061 thousand mu of planned irrigation area. A plurality of natural lakes are distributed along the east line project of the south-to-north water diversion, and the hongze lake, the luoma lake, the south-to-four lake and the east-to-Ping lake are sequentially connected from the river, and the total storage capacity is 45.82 hundred million m³Wherein the upper lakes of the four southern lakes and the east Ping lake do not bear the regulation and storage tasks for a while, after the partial storage capacity is deducted, the total storage capacity is 45.25 hundred million meters³. The water level difference between adjacent lakes is about 10m, if the lakes are taken as nodes, the lakes from the Yangtze river to the lower part of the east-Ping lake can be divided into three large sections, each section is provided with 3 levels of water lifting pump stations, and the total number of the water lifting pump stations is 9.

The method takes the pumping capacity of a pumping station in the open water year as a decision variable, adopts the MQSFLA algorithm to carry out optimization scheduling, and realizes two goals of minimum water shortage and minimum maximum pumping capacity. Through repeated test calculation, the optimal parameter of MQSFLA for solving the multi-objective optimization scheduling problem is determined as：G＝50，N＝10，M＝5，MAXGEN＝5000，K＝10，N_EAThe rotation angle operator delta epsilon (0.001 pi, 0.05 pi), the mutation probability P epsilon (0.1,0.5), and the spatial distribution of the scheduling scheme set are shown in FIG. 4: as can be seen from FIG. 4, the scheduling scheme set appears as a non-convex curve in spatial distribution, the scheduling scheme is widely and uniformly distributed, the two goals of minimum pumping capacity and minimum water shortage are mutually restricted and conflict, and an obvious inverse relationship exists, so that the water resource scheduling scheme set solved by the MQSFLA is reasonable and effective. Based on Pareto optimal solution, determining an optimal reservoir scheduling scheme by adopting a multi-objective decision method, and taking a water shortage subjective weight q₁0.5, subjective weight q of water pumping amount₂0.5; calculating the objective weight p of water shortage₁0.5202 objective weight p of water pumping amount₃0.4498; the preference coefficient l is taken to be 0.5. Determining the minimum water shortage of 21.33 hundred million m³Minimum pumping capacity of 72.63 hundred million m³The optimal dispatching scheme of the lake group, the change diagram of the storage capacity of the east line engineering lake group in south-to-north water diversion is shown in a figure 5, and the water turning amount and the actual water supply condition of the lake entering and the lake exiting in open water are shown in tables 1 and 2.

Table 1 unit of lake inflow and outflow water turnover in open water: hundred million (um)³

Claims

1. A water resource optimization scheduling method based on a multi-target quantum mixed frog leaping algorithm is characterized by comprising the following steps:

the method comprises the following steps: acquiring basic information data of a water resource system project, comprising the following steps: the overflow capacity value q of the pump, the gate station and the reservoir, the initial and final storage capacity limit V of the lake and the reservoir, and the normal water storage level Z of the lake and the reservoir_{Is just}Flood control water level limit Z_DefendDead water level Z_{Death by death}Lake and reservoir volume-water level relation curves S-Z, reservoir downstream water level-discharge flow relation curves Z-Q, reservoir generator setOutput constraint value N and water inflow W;

step two: establishing a target function with maximum comprehensive benefit and the like, and considering constraint conditions of water balance, unit output and overcurrent capacity to obtain a multi-target optimization scheduling mathematical model;

step three: executing a multi-target quantum mixed frog leaping algorithm;

step four: determining an optimal scheduling scheme of a water resource system by adopting a multi-objective decision method based on combined weight based on EA;

(1) determining the water resource scheduling period T by taking the water storage capacity S at the end of the water resource engineering period, the drainage flow u or the engineering water flow q as decision variables;

(2) setting parameters: determining an initial population scale G, a sub-population number N, a sub-population individual number M, a global iteration number MAXGEN, a sub-population iteration number K, and an external archive set scale N which are formed by decision variables_EARotation angle operator [ delta ]_min,δ_max]Frequency of variation [ P ]_min,P_max]；

(4) the initial global iteration times are set, GEN is 0, and an external archive set EA is assigned to be a null set;

(5) performing solution space conversion and calculating each objective function value of each individual, performing non-dominated sorting, updating EA according to a dynamic update mechanism, and randomly selecting a global optimal solution X_g,bDetermining the corresponding amplitude of the optimal solution on the Bloch sphereθ_g,b、The operation flow of the external archive set with the dynamic update mechanism is as follows:

④ reproduction and optimization of Elite individuals in EA, randomly selecting global optimum solution X_g,bDetermining theta_g，b、

① setting the optimal individual X in the current sub-population_bAnd worst individual X_wJ-th qubit, j ═ 1,2, …, T;

2. The water resource optimal scheduling method based on the multi-objective quantum leapfrog algorithm of claim 1, wherein the optimal scheduling scheme of the water resource system is determined by adopting a multi-objective decision-making method based on combination weight based on EA, and the method comprises the following steps:

(1) determining attribute subjectivity using analytic hierarchy processAnd (3) weighting: establishing a hierarchical structure of the system according to the selected target functions, performing pairwise comparison between indexes of all the hierarchies, establishing a judgment matrix, and calculating the weight w 'of each evaluation index'_jJ 1, 2., n, i.e. each evaluation protocol is described by n evaluation indices and a consistency check is performed;

(2) the entropy weight method is used for determining the objective weight of the attributes as follows:

① construct a relative membership matrix EA having N in common_EASet non-inferior solution, i.e. presence of N_EAScheduling scheme of water resource to be evaluated; calculating respective objective function values f of all non-inferior solutions_i,j，i＝1,2,…,N_EAJ 1,2, n, calculating a relative membership matrix R based on index feature matrix normalization_i,j；